Derivation And Validation Of A Risk Scoring System For The Necessity Of Hysteroscopy For Endometrial Lesions

Objective:The purpose of this study is to initially establish and validate a predictive model for the necessity of hysteroscopy to screen low-risk patients to reduce unnecessary hysteroscopy.Methods:Retrospectively collected data from 5,114 cases who had underwent TVS before had hysteroscopy at the Sixth People's Hospital Affiliated to Shanghai Jiao Tong University from January 2007 to July 2019.To observe the incidence and distribution of endometrial diseases treated with hysteroscopy in our hospital,to understand the diagnostic efficacy of hysteroscopy and TVS.Multivariate Logistic regression analysis was used to evaluate the clinical information of patients with endometrial cancer(EC)and endometrial atypical hyperplasia(AEH)and determine the predictors of the necessity of hysteroscopy.These predictors were assigned according to Logistic?regression coefficients to establish a hysteroscopy necessity scoring system.The receiver operator characteristic(ROC)curve and area under the ROC curve(AUC)were used to evaluate the predictive ability of the scoring system,and the Hosmer-Lemeshow test was used to evaluate its calibration ability.Results:The number of patients in our hospital has increased year after year since hysteroscopy and surgery were performed in 2007,especially since 2014,the increase has been larger,and the disease with the highest proportion is endometrial polyps.The diagnostic efficacy of hysteroscopy is significantly higher than TVS(P<0.001).Logistic regression results showed that age,ET thickening,postmenopausal bleeding,low births,and high BMI were common risk factors for patients with EC and AEH.ET value abnormal(ET?8mm in premenopausal cases or ET?4mm in postmenopausal bleeding cases or ET?11mm in postmenopausal no bleeding cases,1points),age?45 years old(1 points),number of births?1(1 points),BMI?24kg/m~2(2 points),TVS indicates that have blood flow signals(2points)and rich blood flow signals(7points),low echo in the uterine cavity or mixed echo(5 points),with postmenopausal bleeding or with abnormal uterine bleeding(17 points)are necessary factors for hysteroscopy.Patients with a score of less than 10 were calculated as low-risk patients.The AUC in the modeling group was 0.918(P<0.001,95%CI:0.908-0.928),the predictive sensitivity was 0.928,and the specificity was 0.908.The AUC in the verification group was 0.931(P<0.001,95%CI:0.914-0.948),the prediction sensitivity was 0.952,and the specificity was 0.910;the overall AUC was 0.921(P<0.001,95%CI:0.912-0.930),The prediction sensitivity is 0.933,and the specificity is 0.909.Conclusion:The hysteroscopy necessity scoring system established in this study based on the clinical characteristics of patients has high predictive sensitivity and specificity.When the method 2 score is less than 10 points,follow-up can be performed,and the score?10 requires hysteroscopy Low-risk patients can be screened to reduce unnecessary invasive procedures.